0
Research Papers

# Inversion-Based Feedforward Approach to Broadband Acoustic Noise Reduction

[+] Author and Article Information
Tom C. Waite

Department of Mechanical Engineering, Iowa State University, 2025 Black Engineering Building, Ames, IA 50011twaite@iastate.edu

Qingze Zou1

Department of Mechanical Engineering, Iowa State University, 2025 Black Engineering Building, Ames, IA 50011qzzou@iastate.edu

Atul Kelkar

Department of Mechanical Engineering, Iowa State University, 2025 Black Engineering Building, Ames, IA 50011akelkar@iastate.edu

1

Corresponding author.

J. Vib. Acoust 130(5), 051010 (Aug 14, 2008) (7 pages) doi:10.1115/1.2948411 History: Received October 03, 2007; Revised March 31, 2008; Published August 14, 2008

## Abstract

In this article, an inversion-based feedforward control approach to achieve broadband active-noise control is investigated. Broadband active-noise control is needed in many areas, from heating, ventilation and air conditioning (HVAC) ducts to aircraft cabins. Achieving broadband active-noise control, however, is very challenging due to issues such as the complexity of acoustic dynamics (which has no natural roll-off at high frequency, and is often nonminimum phase), the wide frequency spectrum of the acoustic noise, and the critical requirement to overcome the delay of the control input relative to the noise signal. These issues have limited the success of existing feedforward control techniques to the low-frequency range of $[0,1]kHz$. The modeling issues in capturing the complex acoustic dynamics coupled with its nonminimum-phase characteristic also prevent the use of high-gain feedback methods, making the design of an effective controller to combat broadband noises challenging. In this article, we explore, through experiments, the potential of inversion-based feedforward control approach for noise control over the $1kHz$ low-frequency range limit. Then we account for the effect of modeling errors on the feedforward input by a recently developed inversion-based iterative control technique. Experimental results presented show that noise reduction of over $10–15dB$ can be achieved in a broad frequency range of $5kHz$ by using the inversion-based feedforward control technique.

<>

## Figures

Figure 7

The comparison of the power spectrum of the residual noise when using the IIC algorithm after the first and the fourth iteration

Figure 6

The comparison of the power spectrum of the noise when using the EIIC algorithm after the first and the fourth iteration

Figure 5

The comparison of the power spectrum, Ψx(jω), with (dashed blue) and without (solid red) applying the inverse-feedforward input uinv(⋅)

Figure 4

The experimentally measured magnitude variation ∣ΔG(jω)∣ of the acoustic dynamics Gsm(jω) (a) and the estimated upper bound of the iterative coefficient ρsup(ω) at each frequency (b)

Figure 3

The experimentally measured frequency response of the dynamics from the control speaker to Mic. 2, Gsm(jω)

Figure 2

The manikin experimental setup for acoustic noise control research; the small picture on the right is the zoomed-in view of the part around the ear indicated by the dashed lines

Figure 1

The block diagram of the system-inversion-based feedforward approach for active-noise control

## Errata

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections